Association between the ERCC5 Asp1104His Polymorphism and Cancer Risk: A Meta-Analysis

Mei-Ling Zhu1,2,3., Mengyun Wang2,3., Zhi-Gang Cao2,4.,JingHe1,2,3, Ting-Yan Shi1,2,3, Kai-Qin Xia1,2,3, Li-Xin Qiu1,2,3*, Qing-Yi Wei3,5* 1 Department of Oncology, Fudan University Shanghai Cancer Center, Shanghai, China, 2 Department of Oncology, Shanghai Medical College, Fudan University, Shanghai, China, 3 Cancer Research Laboratory, Fudan University Shanghai Cancer Center, Shanghai, China, 4 Department of Breast Surgery, Cancer Center and Cancer Institute, Fudan University, Shanghai, China, 5 Department of Epidemiology, The University of Texas MD Anderson Cancer Center, Houston, Texas, United States of America

Abstract

Background: Excision repair cross complementing group 5 (ERCC5 or XPG) plays an important role in regulating DNA excision repair, removal of bulky lesions caused by environmental chemicals or UV light. Mutations in this cause a rare autosomal recessive syndrome, and its functional single nucleotide polymorphisms (SNPs) may alter DNA repair capacity phenotype and cancer risk. However, a series of epidemiological studies on the association between the ERCC5 Asp1104His polymorphism (rs17655, G.C) and cancer susceptibility generated conflicting results.

Methodology/Principal Findings: To derive a more precise estimation of the association between the ERCC5 Asp1104His polymorphism and overall cancer risk, we performed a meta-analysis of 44 published case-control studies, in which a total of 23,490 cases and 27,168 controls were included. To provide additional biological plausibility, we also assessed the genotype-gene expression correlation from the HapMap phase II release 23 data with 270 individuals from 4 ethnic populations. When all studies were pooled, we found no statistical evidence for a significantly increased cancer risk in the recessive genetic models (His/His vs. Asp/Asp: OR = 0.99, 95% CI: 0.92–1.06, P = 0.242 for heterogeneity or His/His vs. Asp/His + Asp/Asp: OR = 0.98, 95% CI: 0.93–1.03, P = 0.260 for heterogeneity), nor in further stratified analyses by cancer type, ethnicity, source of controls and sample size. In the genotype-phenotype correlation analysis from 270 individuals, we consistently found no significant correlation of the Asp1104His polymorphism with ERCC5 mRNA expression.

Conclusions/Significance: This meta-analysis suggests that it is unlikely that the ERCC5 Asp1104His polymorphism may contribute to individual susceptibility to cancer risk.

Citation: Zhu M-L, Wang M, Cao Z-G, He J, Shi T-Y, et al. (2012) Association between the ERCC5 Asp1104His Polymorphism and Cancer Risk: A Meta-Analysis. PLoS ONE 7(7): e36293. doi:10.1371/journal.pone.0036293 Editor: Rui Medeiros, IPO, Inst Port Oncology, Portugal Received January 31, 2012; Accepted March 29, 2012; Published July 18, 2012 Copyright: ß 2012 Zhu et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This study was supported by a grant from ‘‘China’s Thousand Talents Program’’ Recruitment at Fudan University, a grant from the Ministry of Health (201002007) and the National Natural Science Foundation of China (81101808). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * E-mail: [email protected] (L-XQ); [email protected] (Q-YW) . These authors contributed equally to this work.

Introduction in the NER pathway play vital roles in repairing DNA damage and maintain genome integrity [2,3]. Exposure to environmental carcinogens can cause different The excision repair cross complementing group 5 (ERCC5) types of DNA damage that subsequently lead to carcinogenesis of gene, also known as the group G (XPG) different tissues, if left unrepaired. During the evolution, humans gene, is a member of the flap structure-specific endonuclease 1 have developed a versatile DNA repair machinery to ensure (FEN1) family and encodes a of 1186 amino acids. The genome integrity in response to the insults of cancer-causing primary structure of human ERCC5 protein harbors the N- and I- agents. DNA repair is a complex biological process consisting of nuclease domains that are highly conserved, which together form several distinct pathways. To date, more than 150 human DNA the nuclease core [4]. Mutations of several conserved residues in repair have been identified in five major pathways: the active site, including Glu77, Glu791 and Asp812, abolish the nucleotide excision repair (NER), (BER), catalytic activity of the protein [5,6]. N- and I-nuclease domains mismatch repair (MMR), double-strand break repair (DSBR), and are separated by the 600 amino acid spacer region that is highly transcription coupled repair (TCR). Of those pathways, NER is acidic for protein-protein interactions including with TFIIH the most studied DNA repair mechanism responsible for various [7,8,9] and RPA [10] and therefore recruits ERCC5 to the sites types of DNA damage, including thymidine dimers, oxidative of NER [11]. Additionally, ERCC5 contains ubiquitin-binding DNA damage, bulky adducts cross-links, and alkylating damage [1]. At least eight core genes (i.e., ERCC1, XPA, XPB/ERCC3, motif (UBM) as well as a PIP domain that mediates interactions XPC, XPD/ERCC2, XPE/DDB1, XPF/ERCC4, and XPG/ERCC5) with PCNA [12,13]. Such an interaction between ERCC5 and

PLoS ONE | www.plosone.org 1 July 2012 | Volume 7 | Issue 7 | e36293 ERCC5 Polymorphism and Cancer Susceptibility

Figure 1. Structural characteristics and function of ERCC5 protein modified from the picture in the reference [14]. (A) Human ERCC5 protein harbors the N- and I-nuclease domains (blue) and 600 amino acid spacer region (Orange). Conserved residues (Glu77, Glu791 and Asp812) located in the active site (red). Interaction regions with TFIIH, RPA, and PCNA (PIP) and the ubiquitin-binding domain (UBM) are indicated. (B) ERCC5 cleaves 59 flap, splayed-arm and bubble substrates at ss/dsDNA junctions and makes the 39 incision in NER. (C) ERCC5 protein plays versatile roles in cellular processes including DNA repair, genomic integrity maintenance and modulation of gene transcription. doi:10.1371/journal.pone.0036293.g001

PCNA could be involved in triggering the 39 incision in NER. common [minor allele frequency (MAF) .0.05] and regarded as ERCC5 cleaves the 59 flap, splayed arm and a variety of bubble a tagger, which was most frequently investigated for its association substrates at ss/dsDNA junctions with the 59overhang and makes with cancer risk. Interestingly, relatively few nsSNPs are present in the 39 incision in NER [14] (Figure 1). the eight NER core genes, suggesting the conservativeness of these As a structure-specific endonuclease and also a 59-39 exo- genes for their biological importance. nuclease, the ERCC5 protein is required for two sub-pathways in In a published meta-analysis, a total of 12 SNPs of the eight NER. One is TCR, which preferentially removes DNA damage in NER core genes, including 6 nsSNPs, have been investigated for the transcribed DNA strand of active genes; the other is global the associations with cancer risks [29–44], among which 5 nsSNPs genomic repair (GGR), which removes lesions throughout the were found to be associated either with risk of a specific cancer or genome [15,16]. Additionally, ERCC5 also possesses some risk of the overall cancer risks mostly under recessive genetic secondary functions independent of its cleavage activity in models (Table S2), but no published meta-analysis has summa- supporting a role of TFIIH in receptor-mediated transcription rized all reported studies of the Asp1104His polymorphism in [17,18]. Furthermore, data from S. cerevisiae studies demonstrate association with risk of all cancer types. It is biologically plausible a role for Rad2 (the ERCC5 homolog) in RNA polymerase II- that the Asp1104His polymorphism, causing a change from mediated transcription [19]. In addition, ERCC5 is thought to aspartate to histidine at codon 1104 in ERCC5 protein, may result have a possible role in the removal of oxidative damage by BER in an alteration of the gene function, thus likely altering risk of and possibly other pathways [20,21]. Numerous studies using developing cancers, possibly following a recessive genetic model. various tumor cell lines or tissues indicates that ERCC5 plays a key To date, although a number of studies have been performed to role in carcinogenesis and that its deficiency leads to DNA repair investigate the association between the ERCC5 Asp1104His defects, genomic instability, failure of modulation of gene polymorphism and cancer risk, the evidence regarding the role transcription [22–26]. Genetic disorders resulting from mutations of SNPs of the ERCC5 gene as a genetic marker for cancer risk is in the ERCC5 gene, such as xeroderma pigmentosum (XP), conflicting, partially because of the possible lack of a main effect of Cockayne syndrome (CS), and tri-chothiodystrophy (TTD), un- the SNP on risk of any single type of cancer, a possibly low derscore the biological importance of this gene [14], and most of penetrance or weak effect, or the relatively small sample size in these syndromes follow a recessive genetic model, in which each of published studies. Therefore, we performed a meta- heterozygotes are unaffected, but mutant homozygotes manifest analysis to identify statistical evidence for an association between the disease [27]. the ERCC5 Asp1104His polymorphism and cancer risk using all The ERCC5 gene is located on 13q22-q33, consists published data to date. of 15 exons that range from 61 to 1074 bp and 14 introns that range from 250 to 5763 bp, and spans 32 kb [28]. To date, at least 73 non- Materials and Methods synonymous SNPs (nsSNPs) in the ERCC5 coding region have been identified (http://www.ncbi.nlm.nih.gov/SNP/), and 24 SNPs in Identification and Eligibility of Relevant Studies the gene region have been studied for their association with cancer We searched two electronic databases (MEDLINE and risk (Table S1), of which only four were nsSNPs (Figure S1). For EMBASE) for all relevant articles with the following terms: example, the Asp1104His polymorphism (rs17655 G.C) is ‘‘ERCC5’’ or ‘‘XPG’’, ‘‘DNA repair’’, ‘‘polymorphism’’ or ‘‘vari-

PLoS ONE | www.plosone.org 2 July 2012 | Volume 7 | Issue 7 | e36293 ERCC5 Polymorphism and Cancer Susceptibility ant’’, ‘‘case-control’’, ‘‘risk’’, ‘‘association’’, and ‘‘cancer’’ or random effects (DerSimonian and Laird method 1986) [48] and ‘‘carcinoma’’ or ‘‘neoplasm’’ or ‘‘malignance’’ (last search was fixed effects (Mantel–Haenszel method) [49]. When P value of the updated on Sept 1, 2011). References of the retrieved articles or heterogeneity test was .0.10, the fixed-effects model was used, reviews on this topic were also manually screened for additional which indicates no significant heterogeneity of the effect size across relevant eligible studies. all studies; otherwise, the random-effects model was more We defined inclusion criteria as follows: written in English or appropriate, which tends to provide wider CIs, when the results Chinese; case-control design; sufficient information for estimating of the constituent studies differ among themselves. To evaluate the odds ratio (OR) and their 95% confidence interval (CI); observed effect of individual studies on overall risk of cancers, we conducted genotype frequencies in the controls in agreement with Hardy- sensitivity analyses by excluding each study individually and Weinberg equilibrium (HWE). Abstracts and unpublished reports recalculating the ORs and 95% CI. We also used the inverted were not considered. Investigations in subjects with family history funnel plot and the Egger’s test to examine the potential influence or cancer-prone disposition were also excluded. Meanwhile, if of publication bias (linear regression analysis) [50]. HWE among studies had overlapping subjects, we selected the most recent study controls for each study was examined by the Pearson’s goodness- that included the largest number of individuals in the publications. of-fit chi-square test. The boxplot presentation and trend tests Additionally, we also checked for minor allelic frequency (MAF) were performed with Statistical Analysis System software (v.9.1 among studies by different genotype frequencies in ethnic groups SAS Institute, Cary, NC) All other statistical analyses were carried based on Hapmap or dbSNP frequencies reported for the different out with STATA software, version 11.0 (Stata Corporation, ethnic groups, and the datasets were excluded if they had a very College Station, TX). All P values were two-sided with a signifi- high probability of inaccurate reported. cance level of 0.05, unless specified otherwise.

Data Extraction Results Two investigators (Zhu ML and Wang MY) independently extracted the following information from each study: the first Study Characteristics author, year of publication, country of origin, ethnicity, cancer We identified a total of 74 relevant publications after initial type, source of controls (population-based, hospital-based, family- screening. Among these, 62 publications had met the inclusion based and mixed controls), genotyping method, number of criteria and were subjected to further examination. We excluded 8 genotyped cases and controls, numbers of genotypes for ERCC5 publications because they did not present detailed genotyping Asp1104His (rs17655) in cases and controls, and main findings. information [51–58]. We also excluded 3 publications because For studies including subjects of different ethnic groups, we they included the overlapped data with those included in the extracted data separately for each ethnic group and categorized as analysis [59,60,61]. Furthermore, we removed 7 publications Caucasian, Asian, African and others. When a study did not state because their genotype distributions among the controls deviated what ethic groups were included or if it was impossible to separate from HWE [62–68]. Therefore, our final data pooling consisted of participants according to the data presented, we termed the 44 publications [69–112] with a total of 23490 cancer cases and sample as ‘others’. 27168 controls, of which there were actual 49 case-control studies, because 5 publications provided more than one individual study Correlation Analysis of ERCC5 mRNA Expression (Figure S2). These 49 studies included 9 breast cancer studies, 8 We provide biological plausibility of the studied SNP, we skin cancer studies, 5 lung cancer studies, 5 bladder cancer studies, downloaded the Asp1104His genotyping data from the HapMap 3 head and neck cancer studies, 3 colorectal cancer studies, 3 non- phase II release 23 data set consisting of 3.96 million SNP Hodgkin lymphoma studies, and 13 studies of other cancers. Of genotypes from 270 individuals from four populations (CEU: 90 these, there were 27 hospital-based studies, 20 population-based Utah residents from northern and western Europe; CHB: 45 studies, 1 family-based study, and 1 study with mixed controls. In unrelated Han Chinese in Beijing; JPT: 45 unrelated Japanese in addition, 29 of 49 studies were conducted in Caucasians, 4 were Tokyo; YRI: 90 Yoruba in Ibadan, Nigeria) (http://www.sanger. conducted in African-Americans, 6 were conducted in Asians, and ac.uk/humgen/hapmap3) [45]. The data on ERCC5 mRNA the remaining 10 were conducted in other ethnic groups. Several expression levels from EBV-transformed B lymphoblastoid cell genotyping methods were used, including the polymerase chain lines from the same 270 HapMap individuals were available online reaction–restriction fragment length polymorphism (PCR-RFLP), (http://app3.titan.uio.no/biotools/tool.php?app = snpexp) as well which was the most frequently used method, TaqMan, sequenc- [46,47]. Then, we conducted linear regression model-trend test for ing, Illumina, SNaPshot, SNPlex and Mass spectrometry, but two assessing the correlation between Asp1104His and ERCC5 mRNA publications did not provide information about genotyping expression for different populations. methods. Additionally, all studies were in keep with HapMap or dbSNP frequencies reported for the different ethnic groups Statistical Methods (Table S3). We assessed the association between the ERCC5 Asp1104His polymorphism and cancer risk by crude ORs and 95% CIs. Then, Quantitative Synthesis we calculated the pooled ORs and 95% CIs under an assumption of a recessive genetic model (His/His vs. Asp/Asp or His/His vs. When all eligible studies were pooled into one dataset for the Asp/His + Asp/Asp). In addition, we performed stratification meta-analysis, we found no statistical association between the analyses by cancer type (if one cancer type contained less than ERCC5 Asp1104His polymorphism and overall cancer risk under three studies, it was merged into the ‘other cancers’ group), the recessive genetic models: His/His vs. Asp/Asp: OR = 0.99, ethnicity, source of controls, study design and sample size (,500, 95% CI: 0.92–1.06 or His/His vs. Asp/His + Asp/Asp: 500–1000, and .1000). OR = 0.98, 95% CI: 0.93–1.03. We evaluated the between-study heterogeneity by using the Chi In the stratified analysis by ethnicity, we did not observe any square-based Q-test, which was considered significant if P,0.10. association between the polymorphism and cancer risk in the Values from single studies were combined using models of both recessive genetic models, neither and had the similar results in the

PLoS ONE | www.plosone.org 3 July 2012 | Volume 7 | Issue 7 | e36293 ERCC5 Polymorphism and Cancer Susceptibility

Table 1. Meta-analysis of the association between the ERCC5 Asp1104His polymorphism and cancer risk under the XP recessive genetic model for 49 studies.

No. of No. of subjects Variables studies Cases/controls His/His vs. Asp/Asp His/His vs. Asp/His+Asp/Asp c a b c a b OR (95%CI) P OR P het OR (95%CI) P OR P het

All 49 23490/27168 0.99 (0.92–1.06) 0.707 0.242 0.98 (0.93–1.03) 0.393 0.260 Cancer type Breast 9 4915/5277 0.99 (0.85–1.15) 0.882 0.362 0.95 (0.83–1.09) 0.497 0.420 Skin 8 3909/4263 0.95 (0.78–1.16) 0.622 0.586 0.98 (0.86–1.13) 0.809 0.782 Lung 5 4702/5654 0.99 (0.86–1.15) 0.937 0.032 0.98 (0.90–1.06) 0.566 0.049 Bladder 5 2304/2253 0.94 (0.74–1.20) 0.621 0.112 0.95 (0.75–1.21) 0.686 0.083 Head and neck 3 1429/1954 0.90 (0.67–1.22) 0.510 0.194 0.88 (0.67–1.16) 0.364 0.240 Colorectal 3 743/879 0.97 (0.59–1.58) 0.900 0.372 0.92 (0.57–1.48) 0.720 0.262 NHL 3 2105/1957 1.07 (0.83–1.38) 0.594 0.238 1.03 (0.80–1.32) 0.839 0.345 Others 13 3383/4931 1.02 (0.86–1.21) 0.850 0.384 1.03 (0.89–1.19) 0.696 0.244 Ethnicity Caucasian 29 13316/15586 0.99 (0.89–1.10) 0.814 0.287 0.98 (0.90–1.08) 0.709 0.452 African-American 4 1318/1330 1.11 (0.89–1.38) 0.365 0.020 1.03 (0.86–1.24) 0.747 0.016 Asian 6 2314/2443 0.90 (0.76–1.06) 0.216 0.292 0.93 (0.82–1.06) 0.262 0.091 Others 10 6542/7809 1.00 (0.88–1.15) 0.982 0.878 0.98 (0.90–1.07) 0.679 0.852 Source of controls Hospital 27 9787/11586 0.95 (0.85–1.05) 0.298 0.186 0.97 (0.89–1.05) 0.426 0.129 Population 20 10333/11150 1.02 (0.91–1.14) 0.732 0.325 0.99 (0.89–1.09) 0.795 0.427 Others 2 3370/4432 1.02 (0.85–1.23) 0.810 0.853 0.98 (0.90–1.08) 0.706 0.795 Sample size ,500 33 7469/10388 0.98 (0.87–1.11) 0.755 0.055 0.96 (0.86–1.06) 0.396 0.067 500–1000 11 8170/7890 1.02 (0.90–1.15) 0.781 0.704 1.01 (0.91–1.11) 0.923 0.649 .1000 5 7851/8890 0.96 (0.84–1.09) 0.516 0.893 0.97 (0.90–1.05) 0.474 0.916

aP value of the Z-test for odds ration test. bP value of the Q-test for heterogeneity test. cFixed effects model. doi:10.1371/journal.pone.0036293.t001 stratified analyses by tumor type, source of controls, and sample obvious publication bias. Furthermore, the Egger’s test further size in cases (Table 1, Figure 2). provided statistical evidence that there was no significant publication bias in this meta-analysis (the Egger’s test: His/His Heterogeneity and Sensitivity Analyses vs. Asp/Asp: P = 0.897, His/His vs. Asp/His + Asp/Asp: There were no between-study heterogeneity among overall P = 0.749). studies of the ERCC5 Asp1104His polymorphism in the recessive genetic models (x2 = 54.45, P = 0.242 for heterogeneity test and Correlation Analysis for ERCC5 mRNA Expression and 2 x = 53.86, P = 0.260 for heterogeneity test for His/His vs. Asp/ Asp1104His Genotypes Asp and His/His vs. Asp/His + Asp/Asp, respectively). In the In the genotype-phenotype correlation analysis using the sensitivity analyses, the influence of each study on the pooled OR lymphoblastoid cell lines derived from peripheral lymphocytes was checked by repeating the meta-analysis while omitting each from 270 people, we found no trend for the allele effect on ERCC5 study, one at a time. This procedure validated the stability of our mRNA expression for Europeans (P trend = 0.308), Asians (P results. Furthermore, the inclusion of 7 studies, whose genotype trend = 0.091) and Africans (P trend = 0.308) (Figure 3). distributions among the controls deviated from HWE, affected 2 between-study heterogeneity for His/His vs. Asp/Asp (x = 72.21, Discussion P = 0.060), but did not influence the result of the meta-analysis significantly: His/His vs. Asp/Asp: OR = 1.00, 95% CI: 0.93– On the basis of eligible 49 case-control studies with a total of 1.09. His/His vs. Asp/His + Asp/Asp: OR = 0.98, 95% CI: 0.93– 23490 cancer cases and 27168 controls, our meta-analysis 1.03. comprehensively evaluated the association between the ERCC5 Asp1104 His polymorphism and risk of different types of cancers, Publication Bias and we did not find statistical evidence for such an association in We conducted Begg’s funnel plot and Egger’s test to access the the recessive genetic models as shown in the XP syndrome. publication bias of all included studies. The shape of the funnel Similarly, in subgroup analyses, we consistently showed no plot seemed symmetrical (Figure S3), suggesting that there was no statistical association between the polymorphism and cancer risk

PLoS ONE | www.plosone.org 4 July 2012 | Volume 7 | Issue 7 | e36293 ERCC5 Polymorphism and Cancer Susceptibility

Figure 2. Forest plot of overall cancer risk of different cancer types associated with ERCC5 Asp1104His polymorphism (His/His vs. Asp/His + Asp/Asp) by the fixed effects for each of the 44 published studies. For each study, the estimates of OR and its 95% CI were plotted with a box and a horizontal line. The symbol filled diamond indicates pooled OR and its 95% CI. doi:10.1371/journal.pone.0036293.g002

PLoS ONE | www.plosone.org 5 July 2012 | Volume 7 | Issue 7 | e36293 ERCC5 Polymorphism and Cancer Susceptibility

Figure 3. Correlation between Asp1104His and ERCC5 mRNA expression for different populations. A: CEU, 90 Utah residents from northern and western Europe; B: Asians, 45 unrelated Han Chinese in Beijing (CHB) and 45 unrelated Japanese in Tokyo (JPT); C: YRI, 90 Yoruba in Ibadan, Nigeria. doi:10.1371/journal.pone.0036293.g003 in any of the subgroups. Furthermore, the observed null variants are unlikely to have a significant biological effect. For associations were supported by the data that the variant genotypes common variants, in most cases, the disease-associated variant of the SNP were not associated with mRNA expression levels of itself is unlikely to be functionally relevant [115]. The third ERCC5 in the lymphoblastoid cell lines. possibility is that the genetic risk of cancer conferred by the DNA repair deregulation is a crucial factor in the multistep common variants is very modest and the penetrance of the process of carcinogenesis, and the ERCC5 gene is a vital variants is very small, which means that even if the polymorphism component of the DNA repair machinery. It has been observed is crucial for carcinogenesis, extremely large-scale evidence would that deficiency of ERCC5 may result in severe autosomal recessive be necessary to establish with high confidence the presence of diseases including XP, CS and TTD [14] characterized by solar specific associations. The inclusion of rare variants and larger hypersensitivity of the skin, high predisposition for developing samples in future genome-wide association studies would help to cancers (mainly epithelial and melanoma) on areas exposed to reveal low-penetrance susceptibility loci that are more likely to be sunlight. Furthermore, studies have suggested that ERCC5 SNPs associated with cancer risk. are associated with development of some cancers, such as breast Furthermore, we did not observe biological evidence for the cancer [44] and smoking-related cancers [23,24]. These suggest effect of this SNP on the gene expression in terms of mRNA levels, a possible link between the ERCC5 function and development of which biological support for the result of no association. Although cancer. The biological mechanisms of the ERCC5 gene in a -based tool predicted this ERCC5 poly- carcinogenesis may be complicated, among which nsSNPs, leading morphism to be a deleterious substitution [116], and computa- to an amino acid change in the protein product and modulating tional algorithms by SIFT and SNPs3D tools aslo identified the the individual DNA repair capacity phenotype [113,114], may polymorphism with some functional implications (http://compbio. account for some of the known genetic variations related to risk of cs.queensu.ca/F-SNP/), such a potentially functional relevance cancers. However, our meta-analysis suggests that there is no have not been validated experimentally to date. Diversities of statistical evidence for an association between the ERCC5 variant-related risk associations in various kinds of cancer may Asp1104His polymorphism and overall cancer risk, which is result from different mechanisms of carcinogenesis among consistent with the previous two meta-analyses conducted in breast different cancer types. Although some studies have discovered cancer and melanoma, respectively. The former included some some sequence variants in the region of chromosome 5p15.33 and studies departure from HWE in the control population, and the 8q24 that are associated with risk of different cancer types [117– latter only contained three studies. Although we excluded the 122], it is still uncertain whether the same polymorphism may inappropriate studies and expanded the sample size, the null have non-specific effect on different types of cancer. Therefore, results were not altered. Furthermore, as far as our knowledge is further functional studies should be undertaken to explore the concerned, none of the SNPs in NER have ever been identified as mechanism underlying the variant-related associations with cancer susceptibility locus in the published genome-wide association risk. studies (GWAS) for common diseases including cancers based on It would be hard to interpret results, if significant heterogeneity common SNPs, which are similar to our results. This is a challenge were present. However, in this meta-analysis, we did not find any to the theory of common variants and common diseases [115]. It is obvious heterogeneity and publish bias across studies. Neverthe- likely that, as NER genes are considered susceptibility genes, the less, some limitations should be addressed. Firstly, although funnel role of NER variants in cancer development may be dependent on plot and Egger’s test show no publication bias, selection bias could the degree of exposures that cause damage to DNA. Therefore, have occurred because only studies published in English and without detailed information about such exposures for further Chinese were included. Secondly, because the reference groups adjustment or stratification, the results of the observed associations were not uniformly defined, some selected population-based may be either biased or masked. For example, XP patients can controls and some used hospital-based cancer-free controls, non- drastically reduce risk of developing skin cancer by avoiding differential misclassification bias is possible; in addition, some exposure to sunlight. Another possibility is that the common control groups may not be representative of the general

PLoS ONE | www.plosone.org 6 July 2012 | Volume 7 | Issue 7 | e36293 ERCC5 Polymorphism and Cancer Susceptibility population. Thirdly, our results were based on unadjusted OR (TIF) estimates, because not all published studies presented adjusted Figure S2 Flow chart of included studies for this meta- ORs or when they did, the ORs were not adjusted by the same analysis. potential confounders, such as age, sex and exposure variables. (TIF) Thus, more comprehensive individual datasets are needed to allow for the adjustment by some co-variants and further evaluation of Figure S3 Funnel plot analysis to detect publication bias potential gene-environmental interactions for susceptibility to for ERCC5 Asp1104His under the recessive genetic cancer. Fourthly, although the sample size of our study was models (A, His/His vs. Asp/Asp and B, His/His vs. relatively large, the statistical power was still limited in the analyses Asp/His + Asp/Asp) for all 44 studies. Each point of subgroups with small sample sizes, particularly when the effect represents an individual study for the indicated association. size is small. Therefore, studies with larger sample sizes with (TIF) sufficient large subgroups should be undertaken to validate our findings. Table S1 Summary of 24 SNPs of the ERCC5/XPG gene In summary, our meta-analysis shows that the ERCC5 that have been studied for their associations with cancer Asp1104His polymorphism appeared to be unlikely to confer risk. susceptibility to cancers. Further well-designed studies with larger (DOCX) sample sizes will be necessary to validate the findings in the present Table S2 Summary of Studied SNPs in the eight NER meta-analysis. genes reviewed in all published meta-analysis. (DOCX) Supporting Information Table S3 Characteristics of the 44 references included Figure S1 ERCC5 gene map labeled with nine SNPs have in the meta-analysis for ERCC5 Asp1104His. been studied for associations with cancer risk. (A) Six (DOCX) SNPs are located in the coding region, among which four are nsSNPs, of which two are synonymous SNPs; two SNPs are Author Contributions located in the 59 untranslated region, and one SNP is located in the 39 untranslated region; six SNPs are tagging SNPs. (B) Nine Conceived and designed the experiments: L-XQ Q-YW. Analyzed the SNPs with a minor allelic frequency in different populations data: M-LZ M-YW L-XQ. Wrote the paper: M-LZ M-YW JH T-YS K- QX L-XQ Q-YW Z-GC. obtained from the dbSNP database.

References 1. Wood RD (1999) DNA damage recognition during nucleotide excision repair 15. Hanawalt PC (2001) Controlling the efficiency of excision repair. Mutat Res in mammalian cells. Biochimie 81: 39–44. 485: 3–13. 2. Wakasugi M, Reardon JT, Sancar A (1997) The non-catalytic function of XPG 16. Emmert S, Slor H, Busch DB, Batko S, Albert RB, et al. (2002) Relationship of protein during dual incision in human nucleotide excision repair. J Biol Chem neurologic degeneration to genotype in three xeroderma pigmentosum group 272: 16030–16034. G patients. J Invest Dermatol 118: 972–982. 3. O’Donovan A, Davies AA, Moggs JG, West SC, Wood RD (1994) XPG 17. Ito S, Kuraoka I, Chymkowitch P, Compe E, Takedachi A, et al. (2007) XPG endonuclease makes the 39 incision in human DNA nucleotide excision repair. stabilizes TFIIH, allowing transactivation of nuclear receptors: implications for Nature 371: 432–435. Cockayne syndrome in XP-G/CS patients. Mol Cell 26: 231–243. 4. Scherly D, Nouspikel T, Corlet J, Ucla C, Bairoch A, et al. (1993) 18. Scharer OD (2008) Hot topics in DNA repair: the molecular basis for different Complementation of the DNA repair defect in xeroderma pigmentosum disease states caused by mutations in TFIIH and XPG. DNA Repair (Amst) 7: group G cells by a human cDNA related to yeast RAD2. Nature 363: 182–185. 339–344. 5. Wakasugi M, Reardon JT, Sancar A (1997) The Non-catalytic function of XPG 19. Lee SK, Yu SL, Prakash L, Prakash S (2002) Requirement of yeast RAD2, protein during dual incision in human nucleotide excision repair. Journal of a homolog of human XPG gene, for efficient RNA polymerase II transcription. Biological Chemistry 272: 16030–16034. implications for Cockayne syndrome. Cell 109: 823–834. 6. Constantinou A, Gunz D, Evans E, Lalle P, Bates PA, et al. (1999) Conserved 20. Klungland A, Hoss M, Gunz D, Constantinou A, Clarkson SG, et al. (1999) residues of human XPG protein important for nuclease activity and function in Base excision repair of oxidative DNA damage activated by XPG protein. nucleotide excision repair. J Biol Chem 274: 5637–5648. Molecular Cell 3: 33–42. 7. Iyer N, Reagan MS, Wu KJ, Canagarajah B, Friedberg EC (1996) Interactions 21. Bessho T (1999) Nucleotide excision repair 39 endonuclease XPG stimulates involving the human RNA polymerase II transcription/nucleotide excision the activity of base excision repairenzyme thymine glycol DNA glycosylase. repair complex TFIIH, the nucleotide excision repair protein XPG, and Nucleic Acids Res 27: 979–983. Cockayne syndrome group B (CSB) protein. Biochemistry 35: 2157–2167. 22. Koeppel F, Poindessous V, Lazar V, Raymond E, Sarasin A, et al. (2004) 8. Araujo SJ, Nigg EA, Wood RD (2001) Strong functional interactions of TFIIH Irofulven cytotoxicity depends on transcription-coupled nucleotide excision with XPC and XPG in human DNA nucleotide excision repair, without repair and is correlated with XPG expression in solid tumor cells. Clinical a preassembled repairosome. Molecular and Cellular Biology 21: 2281–2291. Cancer Research 10: 5604–5613. 9. Thorel F, Constantinou A, Dunand-Sauthier I, Nouspikel T, Lalle P, et al. 23. Cheng L, Sturgis EM, Eicher SA, Spitz MR, Wei Q (2002) Expression of (2004) Definition of a short region of XPG necessary for TFIIH interaction and nucleotide excision repair genes and the risk for squamous cell carcinoma of the stable recruitment to sites of UV damage. Mol Cell Biol 24: 10670–10680. head and neck. Cancer 94: 393–397. 10. He ZG, Henricksen LA, Wold MS, Ingles CJ (1995) Rpa Involvement in the 24. Cheng L, Spitz MR, Hong WK, Wei Q (2000) Reduced expression levels of Damage-Recognition and Incision Steps of Nucleotide Excision-Repair. nucleotide excision repair genes in lung cancer: a case-control analysis. Nature 374: 566–569. Carcinogenesis 21: 1527–1530. 11. Dunand-Sauthier I, Hohl M, Thorel F, Jaquier-Gubler P, Clarkson SG, et al. 25. Bartolucci R, Wei J, Sanchez JJ, Perez-Roca L, Chaib I, et al. (2009) XPG (2005) The spacer region of XPG mediates recruitment to nucleotide excision mRNA expression levels modulate prognosis in resected non-small-cell lung repair complexes and determines substrate specificity. Journal of Biological cancer in conjunction with BRCA1 and ERCC1 expression. Clin Lung Cancer Chemistry 280: 7030–7037. 10: 47–52. 12. Hofmann K (2009) Ubiquitin-binding domains and their role in the DNA 26. Walsh CS, Ogawa S, Karahashi H, Scoles DR, Pavelka JC, et al. (2008) damage response. DNA Repair 8: 544–556. ERCC5 is a novel biomarker of ovarian cancer prognosis. Journal of Clinical 13. Gary R, Ludwig DL, Cornelius HL, MacInnes MA, Park MS (1997) The DNA Oncology 26: 2952–2958. repair endonuclease XPG binds to proliferating cell nuclear antigen (PCNA) 27. Kraemer KH, Lee MM, Andrews AD, Lambert WC (1994) The role of and shares sequence elements with the PCNA binding regions of FEN-1 and sunlight and DNA repair in melanoma and nonmelanoma skin cancer. The cyclin-dependent kinase inhibitor . Journal of Biological Chemistry 272: xeroderma pigmentosum paradigm. Arch Dermatol 130: 1018–1021. 24522–24529. 28. Emmert S, Schneider TD, Khan SG, Kraemer KH (2001) The human XPG 14. Fagbemi AF, Orelli B, Scharer OD (2011) Regulation of endonuclease activity gene: gene architecture, alternative splicing and single nucleotide polymorph- in human nucleotide excision repair. DNA Repair 10: 722–729. isms. Nucleic Acids Res 29: 1443–1452.

PLoS ONE | www.plosone.org 7 July 2012 | Volume 7 | Issue 7 | e36293 ERCC5 Polymorphism and Cancer Susceptibility

29. Li Y, Gu S, Wu Q, Li Y, Fu X, et al. (2007) No association of ERCC1 C8092A 55. Liu YH, Scheurer ME, El-Zein R, Cao YM, Do KA, et al. (2009) Association and T19007C polymorphisms to cancer risk: a meta-analysis. European journal and Interactions between DNA Repair Gene Polymorphisms and Adult of human genetics : EJHG 15: 967–973. Glioma. Cancer Epidemiology Biomarkers & Prevention 18: 204–214. 30. Cao C, Zhang Y-M, Wang R, Sun S-F, Chen Z-B, et al. (2011) Excision repair 56. Michiels S, Danoy P, Dessen P, Bera A, Boulet T, et al. (2007) Polymorphism cross complementation group 1 polymorphisms and lung cancer risk: a meta- discovery in 62 DNA repair genes and haplotype associations with risks for lung analysis. Chinese Medical Journal 124: 2203–2208. and head and neck cancers. Carcinogenesis 28: 1731–1739. 31. Vineis P, Manuguerra M, Kavvoura FK, Guarrera S, Allione A, et al. (2009) A 57. Mort R, Mo L, McEwan C, Melton DW (2003) Lack of involvement of field synopsis on low-penetrance variants in DNA repair genes and cancer nucleotide excision repair gene polymorphisms in colorectal cancer. British susceptibility. Journal of the National Cancer Institute 101: 24–36. Journal of Cancer 89: 333–337. 32. Kiyohara C, Takayama K, Nakanishi Y (2010) Lung cancer risk and genetic 58. Wen H, Ding Q, Fang ZJ, Xia GW, Fang J (2009) Population study of genetic polymorphisms in DNA repair pathways: a meta-analysis. Journal of nucleic polymorphisms and superficial bladder cancer risk in Han-Chinese smokers in acids 2010: 701760. Shanghai. International Urology and Nephrology 41: 855–864. 33. Qian B, Zhang H, Zhang L, Zhou X, Yu H, et al. (2011) Association of genetic 59. An J, Liu Z, Hu Z, Li G, Wang LE, et al. (2007) Potentially functional single polymorphisms in DNA repair pathway genes with non-small cell lung cancer nucleotide polymorphisms in the core nucleotide excision repair genes and risk risk. Lung cancer (Amsterdam, Netherlands) 73: 138–146. of squamous cell carcinoma of the head and neck. Cancer Epidemiol 34. Qiu L, Wang Z, Shi X, Wang Z (2008) Associations between XPC Biomarkers Prev 16: 1633–1638. polymorphisms and risk of cancers: A meta-analysis. European journal of 60. Weiss JM, Weiss NS, Ulrich CM, Doherty JA, Chen C, et al. (2006) Nucleotide cancer (Oxford, England : 1990) 44: 2241–2253. excision repair genotype and the incidence of endometrial cancer: Effect of 35. Stern MC, Lin J, Figueroa JD, Kelsey KT, Kiltie AE, et al. (2009) other risk factors on the association. Gynecologic Oncology 103: 891–896. Polymorphisms in DNA repair genes, smoking, and bladder cancer risk: 61. Rouissi K, Ouerhani S, Hamrita B, Bougatef K, Marrakchi R, et al. (2011) findings from the international consortium of bladder cancer. Cancer Research Smoking and polymorphisms in xenobiotic metabolism and DNA repair genes 69: 6857–6864. are additive risk factors affecting bladder cancer in Northern Tunisia. 36. Zheng W, Cong X-F, Cai W-H, Yang S, Mao C, et al. (2011) Current Pathology and Oncology Research 17: 879–886. evidences on XPC polymorphisms and breast cancer susceptibility: a meta- 62. Berhane N, Sobti RC, Mahdi SA (2012) DNA repair genes polymorphism analysis. Breast cancer research and treatment 128: 811–815. (XPG and XRCC1) and association of prostate cancer in a north Indian 37. Wang F, Chang D, Hu F-l, Sui H, Han B, et al. (2008) DNA repair gene XPD population. Molecular biology reports 39: 2471–2479. polymorphisms and cancer risk: a meta-analysis based on 56 case-control 63. Goncalves FT, Francisco G, de Souza SP, Luiz OC, Festa-Neto C, et al. (2011) studies. Cancer epidemiology, biomarkers & prevention : a publication of the European ancestry and polymorphisms in DNA repair genes modify the risk of American Association for Cancer Research, cosponsored by the American melanoma: A case-control study in a high UV index region in Brazil. Journal of Society of Preventive Oncology 17: 507–517. Dermatological Science 64: 59–66. 38. Li C, Jiang Z, Liu X (2009) XPD Lys(751)Gln and Asp (312)Asn 64. He X, Ye F, Zhang J, Cheng Q, Shen J, et al. (2008) Susceptibility of XRCC3, polymorphisms and bladder cancer risk: a meta-analysis. Molecular biology XPD, and XPG genetic variants to cervical carcinoma. Pathobiology 75: 356– reports 37: 301–309. 363. 39. Mocellin S, Verdi D, Nitti D (2009) DNA repair gene polymorphisms and risk 65. Jeon HS, Kim KM, Park SH, Lee SY, Choi JE, et al. (2003) Relationship of cutaneous melanoma: a systematic review and meta-analysis. Carcinogenesis between XPG codon 1104 polymorphism and risk of primary lung cancer. 30: 1735–1743. Carcinogenesis 24: 1677–1681. 40. Pabalan N, Francisco-Pabalan O, Sung L, Jarjanazi H, Ozcelik H (2010 ) 66. McKean-Cowdin R, Barnholtz-Sloan J, Inskip PD, Ruder AM, Butler M, et al. Meta-analysis of two ERCC2 (XPD) polymorphisms, Asp312Asn and (2009) Associations between Polymorphisms in DNA Repair Genes and Lys751Gln, in breast cancer. Breast cancer research and treatment 124: Glioblastoma. Cancer Epidemiology Biomarkers & Prevention 18: 1118–1126. 531–541. 67. Smith TR, Levine EA, Freimanis RI, Akman SA, Allen GO, et al. (2008) 41. Qiu L-X, Yao L, Zhang J, Zhu X-D, Zhao X-M, et al. (2010) XPD Lys751Gln Polygenic model of DNA repair genetic polymorphisms in human breast polymorphism and breast cancer susceptibility: a meta-analysis involving cancer risk. Carcinogenesis 29: 2132–2138. 28,709 subjects. Breast cancer research and treatment 124: 229–235. 68. Wen SX, Tang PZ, Zhang XM, Zhao D, Guo YL, et al. (2006) Association 42. Chen B, Zhou Y, Yang P, Wu X-T (2011) ERCC2 Lys751Gln and Asp312Asn between genetic polymorphism in xeroderma pigmentosum G gene and risks of polymorphisms and gastric cancer risk: a meta-analysis. Journal of cancer laryngeal and hypopharyngeal carcinomas. Acta Academiae Medicinae Sinicae research and clinical oncology 137: 939–946. 28: 703–706. 43. Yuan L, Cui D, Zhao E-J, Jia C-Z, Wang L-D, et al. (2011) XPD Lys751Gln 69. Kumar R, Hoglund L, Zhao C, Forsti A, Snellman E, et al. (2003) Single polymorphism and esophageal cancer risk: a meta-analysis involving 2288 cases nucleotide polymorphisms in the XPG gene: determination of role in DNA and 4096 controls. World journal of gastroenterology : WJG 17: 2343–2348. repair and breast cancer risk. Int J Cancer 103: 671–675. 44. Ding D-P, He X-F, Zhang Y (2011 ) Lack of association between XPG 70. Mechanic LE, Millikan RC, Player J, de Cotret AR, Winkel S, et al. (2006) Asp1104His and XPF Arg415Gln polymorphism and breast cancer risk: Polymorphisms in nucleotide excision repair genes, smoking and breast cancer a meta-analysis of case-control studies. Breast cancer research and treatment in African Americans and whites: a population-based case-control study. 129: 203–209. Carcinogenesis 27: 1377–1385. 45. The International HapMap Consortium (2003) The International HapMap 71. Shen J, Desai M, Agrawal M, Kennedy DO, Senie RT, et al. (2006) Project. Nature 426: 789–796. Polymorphisms in nucleotide excision repair genes and DNA repair capacity 46. Holm K, Melum E, Franke A, Karlsen TH (2010) SNPexp - A web tool for phenotype in sisters discordant for breast cancer. Cancer Epidemiology calculating and visualizing correlation between HapMap genotypes and gene Biomarkers and Prevention 15: 1614–1619. expression levels. BMC bioinformatics 11: 600–604. 72. Crew KD, Gammon MD, Terry MB, Fang FZ, Zablotska LB, et al. (2007) 47. Stranger BE, Forrest MS, Dunning M, Ingle CE, Beazlsy C, et al. (2007) Polymorphisms in nucleotide excision repair genes, polycyclic aromatic Relative impact of nucleotide and copy number variation on gene phenotypes. hydrocarbon-DNA adducts, and breast cancer risk. Cancer Epidemiology Science 315: 848–853. Biomarkers and Prevention 16: 2033–2041. 48. DerSimonian R, Laird N (1986) Meta-analysis in clinical trials. Control Clin 73. Jorgensen TJ, Visvanathan K, Ruczinski I, Thuita L, Hoffman S, et al. (2007) Trials 7: 177–188. Breast cancer risk is not associated with polymorphic forms of xeroderma 49. Mantel N, Haenszel W (1959) Statistical aspects of the analysis of data from pigmentosum genes in a cohort of women from Washington County, retrospective studies of disease. J Natl Cancer Inst 22: 719–748. Maryland. Breast Cancer Res Treat 101: 65–71. 50. Egger M, Smith GD, Schneider M, Minder C (1997) Bias in meta-analysis 74. Smith TR, Levine EA, Freimanis RI, Akman SA, Allen GO, et al. (2008) detected by a simple, graphical test. British Medical Journal 315: 629–634. Polygenic model of DNA repair genetic polymorphisms in human breast 51. Chen M, Kamat AM, Huang M, Grossman HB, Dinney CP, et al. (2007) cancer risk. Carcinogenesis 29: 2132–2138. High-order interactions among genetic polymorphisms in nucleotide excision 75. Rajaraman P, Bhatti P, Doody MM, Simon SL, Weinstock RM, et al. (2008) repair pathway genes and smoking in modulating bladder cancer risk. Nucleotide excision repair polymorphisms may modify ionizing radiation- Carcinogenesis 28: 2160–2165. related breast cancer risk in US radiologic technologists. International Journal 52. Cui Y, Morgenstern H, Greenland S, Tashkin DP, Mao J, et al. (2006) of Cancer 123: 2713–2716. Polymorphism of Xeroderma Pigmentosum group G and the risk of lung 76. Hsu MS, Yu JC, Wang HW, Chen ST, Hsiung CN, et al. (2010) Synergistic cancer and squamous cell carcinomas of the oropharynx, larynx and Effects of Polymorphisms in DNA Repair Genes and Endogenous Estrogen esophagus. Int J Cancer 118: 714–720. Exposure on Female Breast Cancer Risk. Annals of Surgical Oncology 17: 53. Hung RJ, Baragatti M, Thomas D, McKay J, Szeszenia-Dabrowska N, et al. 760–771. (2007) Inherited predisposition of lung cancer: A hierarchical modeling 77. Blankenburg S, Konig IR, Moessner R, Laspe P, Thoms KM, et al. (2005) No approach to DNA repair and cell cycle control pathways. Cancer Epidemiology association between three xeroderma pigmentosum group C and one group G Biomarkers & Prevention 16: 2736–2744. gene polymorphisms and risk of cutaneous melanoma. Eur J Hum Genet 13: 54. Joshi AD, Corral R, Siegmund KD, Haile RW, Le Marchand L, et al. (2009) 253–255. Red meat and poultry intake, polymorphisms in the nucleotide excision repair 78. Li C, Hu Z, Liu Z, Wang LE, Strom SS, et al. (2006) Polymorphisms in the and mismatch repair pathways and colorectal cancer risk. Carcinogenesis 30: DNA repair genes XPC, XPD, and XPG and risk of cutaneous melanoma: 472–479. a case-control analysis. Cancer Epidemiol Biomarkers Prev 15: 2526–2532.

PLoS ONE | www.plosone.org 8 July 2012 | Volume 7 | Issue 7 | e36293 ERCC5 Polymorphism and Cancer Susceptibility

79. Povey JE, Darakhshan F, Robertson K, Bisset Y, Mekky M, et al. (2007) DNA 99. Shen M, Zheng T, Lan Q, Zhang Y, Zahm SH, et al. (2006) Polymorphisms in repair gene polymorphisms and genetic predisposition to cutaneous melanoma. DNA repair genes and risk of non-Hodgkin lymphoma among women in Carcinogenesis 28: 1087–1093. Connecticut. Hum Genet 119: 659–668. 80. Figl A, Scherer D, Nagore E, Bermejo JL, Botella-Estrada R, et al. (2010) 100. Hill DA, Wang SS, Cerhan JR, Davis S, Cozen W, et al. (2006) Risk of non- Single-nucleotide polymorphisms in DNA-repair genes and cutaneous Hodgkin lymphoma (NHL) in relation to germline variation in DNA repair and melanoma. Mutation Research - Genetic Toxicology and Environmental related genes. Blood 108: 3161–3167. Mutagenesis 702: 8–16. 101. Shen M, Purdue MP, Kricker A, Lan Q, Grulich AE, et al. (2007) 81. Ibarrola-Villava M, Pena-Chilet M, Fernandez LP, Aviles JA, Mayor M, et al. Polymorphisms in DNA repair genes and risk of non-Hodgkin’s lymphoma (2011) Genetic polymorphisms in DNA repair and oxidative stress pathways in New South Wales, Australia. Haematologica 92: 1180–1185. associated with malignant melanoma susceptibility. European Journal of 102. Hussain SK, Mu LN, Cai L, Chang SC, Park SL, et al. (2009) Genetic Cancer. variation in immune regulation and DNA repair pathways and stomach cancer 82. Thirumaran RK, Bermejo JL, Rudnai P, Gurzau E, Koppova K, et al. (2006) in China. Cancer Epidemiol Biomarkers Prev 18: 2304–2309. Single nucleotide polymorphisms in DNA repair genes and basal cell 103. Canbay E, Agachan B, Gulluoglu M, Isbir T, Balik E, et al. (2010) Possible carcinoma of skin. Carcinogenesis 27: 1676–1681. associations of APE1 polymorphism with susceptibility and HOGG1 poly- 83. Wang LE, Li C, Strom SS, Goldberg LH, Brewster A, et al. (2007) Repair morphism with prognosis in gastric cancer. Anticancer Res 30: 1359–1364. capacity for UV light induced DNA damage associated with risk of 104. Weiss JM, Weiss NS, Ulrich CM, Doherty JA, Voigt LF, et al. (2005) nonmelanoma skin cancer and tumor progression. Clin Cancer Res 13: Interindividual variation in nucleotide excision repair genes and risk of 6532–6539. endometrial cancer. Cancer Epidemiol Biomarkers Prev 14: 2524–2530. 84. Sanyal S, Festa F, Sakano S, Zhang Z, Steineck G, et al. (2004) Polymorphisms 105. Doherty JA, Weiss NS, Fish S, Fan W, Loomis MM, et al. (2011) in DNA repair and metabolic genes in bladder cancer. Carcinogenesis 25: 729– Polymorphisms in Nucleotide Excision Repair Genes and Endometrial Cancer 734. Risk. Cancer Epidemiol Biomarkers Prev. 85. Garcia-Closas M, Malats N, Real FX, Welch R, Kogevinas M, et al. (2006) 106. Hooker S, Bonilla C, Akereyeni F, Ahaghotu C, Kittles RA (2008) NAT2 and Genetic variation in the nucleotide excision repair pathway and bladder cancer NER genetic variants and sporadic prostate cancer susceptibility in African risk. Cancer Epidemiol Biomarkers Prev 15: 536–542. Americans. Prostate Cancer Prostatic Dis 11: 349–356. 86. Wu X, Gu J, Grossman HB, Amos CI, Etzel C, et al. (2006) Bladder cancer 107. Le Morvan V, Longy M, Bonaiti-Pellie C, Bui B, Houede N, et al. (2006) predisposition: a multigenic approach to DNA-repair and cell-cycle-control Genetic polymorphisms of the XPG and XPD nucleotide excision repair genes genes. Am J Hum Genet 78: 464–479. in sarcoma patients. Int J Cancer 119: 1732–1735. 87. Narter KF, Ergen A, Agachan B, Gormus U, Timirci O, et al. (2009) Bladder 108. Biason P, Hattinger CM, Innocenti F, Talamini R, Alberghini M, et al. (2011) cancer and polymorphisms of DNA repair genes (XRCC1, XRCC3, XPD, Nucleotide excision repair gene variants and association with survival in XPG, APE1, hOGG1). Anticancer Res 29: 1389–1393. osteosarcoma patients treated with neoadjuvant chemotherapy. 88. Rouissi K, Ouerhani S, Hamrita B, Bougatef K, Marrakchi R, et al. (2011) Pharmacogenomics J. Smoking and Polymorphisms in Xenobiotic Metabolism and DNA Repair 109. El-Zein R, Monroy CM, Etzel CJ, Cortes AC, Xing Y, et al. (2009) Genetic Genes are Additive Risk Factors Affecting Bladder Cancer in Northern polymorphisms in DNA repair genes as modulators of Hodgkin disease risk. Tunisia. Pathol Oncol Res. Cancer 115: 1651–1659. 89. Shen M, Berndt SI, Rothman N, Demarini DM, Mumford JL, et al. (2005) 110. Rajaraman P, Hutchinson A, Wichner S, Black PM, Fine HA, et al. (2010) Polymorphisms in the DNA nucleotide excision repair genes and lung cancer DNA repair gene polymorphisms and risk of adult meningioma, glioma, and risk in Xuan Wei, China. Int J Cancer 116: 768–773. acoustic neuroma. Neuro Oncol 12: 37–48. 90. Sakiyama T, Kohno T, Mimaki S, Ohta T, Yanagitani N, et al. (2005) 111. Pan J, Lin J, Izzo JG, Liu Y, Xing J, et al. (2009) Genetic susceptibility to Association of amino acid substitution polymorphisms in DNA repair genes esophageal cancer: the role of the nucleotide excision repair pathway TP53, POLI, REV1 and LIG4 with lung cancer risk. Int J Cancer 114: 730– Carcinogenesis 30: 785–792. 737. 91. Hung RJ, Christiani DC, Risch A, Popanda O, Haugen A, et al. (2008) 112. Li LM, Zeng XY, Ji L, Fan XJ, Li YQ, et al. (2010) [Association of XPC and International Lung Cancer Consortium: pooled analysis of sequence variants in XPG polymorphisms with the risk of hepatocellular carcinoma]. Zhonghua DNA repair and cell cycle pathways. Cancer Epidemiol Biomarkers Prev 17: Gan Zang Bing Za Zhi 18: 271–275. 3081–3089. 113. Vodicka P, Stetina R, Polakova V, Tulupova E, Naccarati A, et al. (2007) 92. Chang JS, Wrensch MR, Hansen HM, Sison JD, Aldrich MC, et al. (2008) Association of DNA repair polymorphisms with DNA repair functional Nucleotide excision repair genes and risk of lung cancer among San Francisco outcomes in healthy human subjects. Carcinogenesis 28: 657–664. Bay Area Latinos and African Americans. Int J Cancer 123: 2095–2104. 114. Friedberg EC (2003) DNA damage and repair. Nature 421: 436–440. 93. Sugimura T, Kumimoto H, Tohnai I, Fukui T, Matsuo K, et al. (2006) Gene- 115. Bodmer W, Bonilla C (2008) Common and rare variants in multifactorial environment interaction involved in oral carcinogenesis: Molecular epidemi- susceptibility to common diseases. Nat Genet 40: 695–701. ological study for metabolic and DNA repair gene polymorphisms. Journal of 116. Ng PC, Henikoff S (2001) Predicting deleterious amino acid substitutions. Oral Pathology and Medicine 35: 11–18. Genome Res 11: 863–874. 94. Abbasi R, Ramroth H, Becher H, Dietz A, Schmezer P, et al. (2009) Laryngeal 117. Rafnar T, Sulem P, Stacey SN, Geller F, Gudmundsson J, et al. (2009) cancer risk associated with smoking and alcohol consumption is modified by Sequence variants at the TERT-CLPTM1L locus associate with many cancer genetic polymorphisms in ERCC5, ERCC6 and RAD23B but not by types. Nature genetics 41: 221–227. polymorphisms in five other nucleotide excision repair genes. Int J Cancer 118. Gudmundsson J, Sulem P, Manolescu A, Amundadottir LT, Gudbjartsson D, 125: 1431–1439. et al. (2007) Genome-wide association study identifies a second prostate cancer 95. Hongxia M, Hongping Y, Zhensheng L, Li EW, Erich MS, et al. (2011) susceptibility variant at 8q24. Nature genetics 39: 631–637. Polymorphisms of XPG/ERCC5 and risk of squamous cell carcinoma of the 119. Yeager M, Orr N, Hayes RB, Jacobs KB, Kraft P, et al. (2007) Genome-wide head and neck. Pharmacogenetics and Genomics. association study of prostate cancer identifies a second risk locus at 8q24. 96. Pardini B, Naccarati A, Novotny J, Smerhovsky Z, Vodickova L, et al. (2008) Nature genetics 39: 645–649. DNA repair genetic polymorphisms and risk of colorectal cancer in the Czech 120. Haiman CA, Patterson N, Freedman ML, Myers SR, Pike MC, et al. (2007) Republic. Mutat Res 638: 146–153. Multiple regions within 8q24 independently affect risk for prostate cancer. 97. Gil J, Ramsey D, Stembalska A, Karpinski P, Pesz KA, et al. (2011) The C/A Nature genetics 39: 638–644. polymorphism in intron 11 of the XPC gene plays a crucial role in the 121. Tomlinson I, Webb E, Carvajal-Carmona L, Broderick P, Kemp Z, et al. modulation of an individual’s susceptibility to sporadic colorectal cancer. Mol (2007) A genome-wide association scan of tag SNPs identifies a susceptibility Biol Rep. variant for colorectal cancer at 8q24.21. Nature genetics 39: 984–988. 98. Canbay E, Cakmakoglu B, Zeybek U, Sozen S, Cacina C, et al. (2011) 122. Kiemeney LA, Thorlacius S, Sulem P, Geller F, Aben KKH, et al. (2008) Association of APE1 and hOGG1 polymorphisms with colorectal cancer risk in Sequence variant on 8q24 confers susceptibility to urinary bladder cancer. a Turkish population. Curr Med Res Opin 27: 1295–1302. Nature genetics 40: 1307–1312.

PLoS ONE | www.plosone.org 9 July 2012 | Volume 7 | Issue 7 | e36293